Post on 10-Feb-2021
transcript
1
Pattern formation by Staphylococcus epidermidis via
droplet evaporation on micropillars arrays at a
surface
A. Susarrey-Arce*, 1, A. Marin2, A. Massey1, A. Oknianska1, Y. Díaz-Fernandez1, J. F.
Hernández-Sánchez3, E. Griffiths1, J. G. E. Gardeniers4, J. H. Snoeijer3, 5 Detlef Lohse3 and R.
Raval*, 1
1Open Innovation Hub for Antimicrobial Surfaces at the Surface Science Research Centre and Department of
Chemistry, University of Liverpool, Oxford Street, UK L69 3BX, Liverpool
2Institute of Fluid Mechanics and Aerodynamics, Bundeswehr University Munich, Germany
3Physics of Fluids Group, MESA+ Institute for Nanotechnology, J. M. Burgers Centre for Fluid Dynamics,
University of Twente, P.O. Box 217, 7500AE Enschede, The Netherlands
4Mesoscale Chemical Systems, MESA+ Institute for Nanotechnology, University of Twente, P.O. Box 217,
7500AE Enschede, The Netherlands
5Mesoscopic Transport Phenomena, Eindhoven University of Technology, Den Dolech 2, 5612 AZ Eindhoven,
The Netherlands
Corresponding author(s):
Prof. Rasmita Raval, Phone: +44 151 794 6981, e-mail: R.Raval@liverpool.ac.uk
Dr. Arturo Susarrey-Arce, Phone: +44 151 794 3541, e-mail: A.Susarrey-Arce@liverpool.ac.uk
Keywords: zipping-wetting, coffee-stain, bacterial viability, microstructures, wetting
mailto:R.Raval@liverpool.ac.ukmailto:A.Susarrey-Arce@liverpool.ac.uk
2
Abstract
We evaluate the effect of epoxy surface structuring on the evaporation of water droplets
containing Staphylococcus epidermidis (S. epidermidis). During evaporation, droplets with S.
epidermidis cells yield to complex wetting patterns such as the zipping-wetting1-3 and coffee-
stain effects. Depending on the height of the microstructure, the wetting fronts propagate
circularly or in a stepwise manner, leading to the formation of octagonal or square-shaped
deposition patterns.4, 5 We observed that the shape of the dried droplets has considerable
influence on the local spatial distribution of S. epidermidis deposited between micropillars.
These changes are attributed to an unexplored interplay between the zipping-wetting1 and the
coffee-stain6 effects in polygonally-shaped droplets containing S. epidermidis. Induced
capillary flows during evaporation of S. epidermidis are modeled with polystyrene particles.
Bacterial viability measurements for S. epidermidis shows high viability of planktonic cells,
but low biomass deposition on the microstructured surfaces. Our findings provide insights into
design criteria for the development of microstructured surfaces on which bacterial propagation
could be controlled, limiting the use of biocides.
3
1. Introduction
The production of biological and chemical materials7, 8 that control the growth and survival
rate of microorganisms9 at surfaces is of great interest for future antimicrobial strategies.10 An
important factor affecting the development of bacterial colonies is the initial adhesion to the
surface, which initiates proliferation and biofilm formation and has major impact in
contamination of medical devices.11-14 For example, S. epidermidis infections can commence
with the introduction of bacteria transferred from the skin during medical device insertion, and
account for at least 22% of bloodstream infections in intensive care unit patients.15 It has
recently been found that a surface with micro(nano)-topography in contact with
microorganisms can influence microbial growth, attachment, and distribution.16, 17 In addition,
modifying surface topography can also create water repellent substrates, which may prevent
infections by reducing bacterial growth and propagation after the evaporation of the liquid.10,
18-22 However, droplets in such superhydrophobic or hydrophobic states are energetically
unstable and eventually the droplet gets impaled by the microscopic structure, losing the
hydrophobic character23-26 and causing the liquid to infiltrate the structure. Such a transition
can however be avoided with suitable engineered micropatterned substrates27-29 with sharp-
edged pillars30-32 or with relatively high microstructures.33-35 In addition, the spreading of the
liquid front is also affected by the pillar geometry, leading to a droplet footprint with a
polygonal shape. This phenomenon has been termed zipping-wetting and it has been observed
for submillimetric-structures.4, 5 As well as forming elaborately patterned footprints on
surfaces,36-40 the dried pattern can have profound effect on the distribution and survival rate of
bacteria on a substrate. However, little is known about how the presence of bacteria in droplets
affects the drying on microstructured surfaces and how the bacterial interaction at the wetting
front affects the resulting bacterial deposition over the substrate. This problem can be compared
to the behavior of particle suspension droplets, which, upon evaporation, have been shown to
4
leave distinct ring-shaped marks on the surface. This phenomenon is known as coffee-stain
effect whereby the colloidal particles are deposited around the perimeter of a droplet.6, 41
In this paper we assess the effect of epoxy surface structuring on the evaporation of bacteria-
containing droplets and the resulting bacterial distribution on the microstructured surfaces.
First, S. epidermidis wetting patterns are studied. Our experiments show an interesting
combination of the zipping-wetting and the coffee-stain effect that has not been previously
explored for bacterial-containing droplets. The combination of these two phenomena leads to
a breakdown of the droplets axial symmetry which directs the distribution of bacteria along
and outside the droplet perimeter. Second, the local distribution of S. epidermidis cells
deposited between individual micropillars is studied. Our results reveal that the proportion of
the resulting local bacterial patterns can be modified by varying the pillar height of the
fabricated microstructures. Third, S. epidermidis viability is studied and shows that in spite of
high viability of planktonic cells re-grown over the substrates, biofilm formation over these
surfaces is relatively impaired. These effects could be attributed to the local bacterial
distribution over microstructured substrates. Finally, to quantify the dynamics of the S.
epidermidis deposition, polystyrene (PS) particles are used. PS particles resembled the
capillary driven flows during the zipping-wetting and the coffee-stain effects.
5
2. Methods
2.1 Deep Reactive-Ion Etching of Silicon. Photolithographically defined silicon micropillar
arrays were produced with Deep Reactive Ion beam Etching (DRIE) as described in detail
elsewhere.27, 31 In a DRIE system (Adixen AMS100-SE), with a RF generator at 13.56 MHz,
CCP 80 W LF and 1500 W ICP plasma source, the micropillar arrays were etched by keeping
the total chamber pressure at 75 mTorr. The temperature of the electrode with the silicon
substrate was kept at 10 °C, using liquid nitrogen as a coolant. The etching time was varied
from 1.5 min to 5 min to obtain pillar heights of approximately 5 (H5), 10 (H10) and 15 (H15)
μm. SF6 and C4F8 flows were kept constant during the etching process at 250 sccm (standard
cubic centimeter per minute) and 200 sccm, respectively. After the silicon etching, photoresist
and fluorocarbons were stripped in O2 plasma at 500 W for 30 min, a subsequent 1% HF
treatment was used to remove formed SiO2.
2.2 Fabrication of polydimethylsiloxane (PDMS) molds. Prior to the fabrication of PDMS
molds, vapor deposition of trichloro (1H, 1H, 2H, 2H-perfluorooctyl) silane (FOTS from
Fluorochem) was carried out in a vacuum system for 3 min. A negative replica of the pillar
substrate was produced by casting PDMS (Dow Sylgard 184 Silicon elastomer) onto the silicon
etched substrate described in section 2.1. To cure the PDMS, a 1:10 ratio of the curing agent
and the pre-polymer were mixed, degassed and incubated at 85 oC for 3 h. The PDMS mold
was removed from the silicon substrate and cut prior to use. The PDMS mold was then cleaned
extensively with ethanol and isopropanol, dried and treated in air plasma for 1 min in a Femto
Diener plasma cleaner (Zepto model).
2.3 Fabrication of epoxy micropillars. Epoxy micropillars were produced by casting EPO-
TEK (OG142-13 from Epoxy Technology) onto the negative PDMS replica described in
section 2.2. After Epoxy was cast, a glass slide was placed over the PDMS substrate with
6
Epoxy material. The epoxy was cured using Ultraviolet light. A UVL-56 Handheld UV lamp
was used (6 Watt and wavelength of 365 nm) for 30 min followed by incubation at 30 oC for
30 s.
2.4 Configuration of micropillars on epoxy substrates. Epoxy micropillars were fabricated
by casting and curing epoxy glue on a negative PDMS micropillar-replica as described in
section 2.3. These microstructures, labelled from (a) to (c), are shown in Figure 1. The
diameter (d) and interspacing (i) were restricted in the range presented in Table 1, but the
heights (h) were varied from 5 to 15 µm. The configuration of the microstructures is in a square
lattice with a periodicity p = i + d with a packing fraction Φ, calculated as (π/4)(d/p)2 of about
0.19 and aspect ratios (h/d) of approximately 1, 2, and 3 for (a), (b), and (c), respectively. The
outside walls of the micropillars are smooth at the micron-scale for all of the substrates.
2.5 Determination of S. epidermidis cell viability after evaporation of bacterial suspension
over structured surfaces. S. epidermidis (ATTC-12228) cultures were grown over night (200
rpm, at 37 oC) in nutrient broth (NB) medium (Oxoid, Ltd-Thermo Fisher). The bacterial cells
were adjusted to 6.3x106, 8.0x107 and 5.0x109 colony forming units per milliliter (CFU/mL) in
sterile deionized water.
S. epidermidis viability was carried out with Flat and structured epoxy micropillar substrates
sterilized under UV light for 20 min. 10 µL droplets of fresh bacterial cell suspension (9x107
CFU/mL in water) were deposited onto H5, H10, H15 and flat surfaces until complete
evaporation for 30 min. After complete evaporation, each substrate was rehydrated in 1 mL of
NB and the cells were cultured for 24 h at 37 oC. Counting of viable cells was performed after
washing the surface with 200 µL of sterile phosphate-buffered saline (PBS) and serial dilutions.
The experiments were performed in triplicates.
7
2.6 S. epidermidis biofilm formation assay. Microtiter plate biofilm formation assay was
modified from the method described by O’Toole et al.42 Briefly, S. epidermidis cultures were
grown overnight (200 rpm, at 37 oC) in NB medium (Oxoid, Ltd-Thermo Fisher) and diluted
to 107 CFU/mL in NB. Polystyrene flat (PSflat), flat epoxy, and epoxy micropillar substrates
(H5, H10, and H15) of 1 cm x 1cm were sterilized under UV light for 20 min. The substrates
were placed in wells of the 24 well microtiter plate, covered with 600 µL of S. epidermidis 107
cell suspension and incubated for 24h at 37 oC. After incubation bacterial cell suspension was
removed, materials were gently washed 5 times with PBS, moved to the new plate and dried.
The biofilms formed were stained with 600 µL of a 0.1% crystal violet for 15 min at room
temperature (RT). Crystal violet was removed; materials were washed 5 times with sterile
water and dried. For quantification of biofilms formed on the flat and structured substrates, 500
µL of absolute ethanol was added (for 15 min at RT) to solubilize the stain and transferred to
a new plate. The optical density (O.D.) 595 nm was measured in a UV/VIS plate reader
(FilterMax F5 Multi Mode Microplate Reader, Molecular Devices). Three independent
experiments were performed.
2.7 Contact angle measurements on Epoxy micropillar arrays. Contact angle measurements
were performed by placing a water droplet of 2-4 μL on the Epoxy substrates with the set-up
presented in Figure SI-1. Evaporation occurred at room temperature (21° ± 3 °C) in an
atmosphere with a relative humidity of 35±5%. The water was purified in a Millipore Milli-Q
system which involves reverse osmosis, ion-exchange, and filtration steps (18.6 MΩ cm). Side-
view videos were captured via a CMOS camera equipped with x5-x40 magnifying lenses and
with a recording time of 1-2 fps.
Contact angle measurements of water and S. epidermidis droplets on epoxy surfaces were
carried out by placing a water droplet with bacteria suspension of 6.3x106 CFU/mL, 8.0x107
CFU/mL and 5.0x109 CFU/mL on the epoxy substrates. After deposition, the droplets
8
evaporated at room temperature. Top-view droplet evaporation images were recorded at frame
rates of 10 fps with a camera (Photron Fastcam SA7) with a 50D-20x-VI lens mounted in a
Nikon light-microscope. Under such conditions, 2-4 μL droplets evaporate completely in
approximately 1200 s ± 250 s. Contact angle (CA) measurements as a function of time are
shown in Figure SI-2.
2.8 Deposition of Polystyrene particles on Epoxy substrates. A 107 particles/mL solution of
FluoRed-polystyrene (PS) particles purchased from Microparticles GmbH with mean diameter
of 1.2 µm ± 0.04 µm was prepared with deionized water (Milli-Q). Droplets of 2-4 µL were
deposited on the epoxy substrates. Substrate inspection was performed with an inverted
microscope illuminated with a continuous solid-state laser diode pumped at 100 mW (or a
halogen light) to avoid overheating. The images were collected with a CCD camera PCO
Sensicam at 1 frames per second (fps). The droplets were evaporated at 23 oC and 40% relative
humidity. Under such conditions, a 2-4 µL droplet completely evaporated in approximately
1200 s ± 250 s. It is important to note that static contact angle of the droplets containing PS
particles over substrates were very similar, all being slightly below 100˚.
2.9 SEM and AFM characterization. Fracturing the epoxy/glass substrates with a diamond
cutter, a cross sectional scanning electron microscopy (SEM) image of the fabricated epoxy
micropillars was collected with accelerating voltages of 3 kV and x1,300 magnification using
a JSM-6610 JEOL SEM. To increase the electrical conductivity of the micropillars, prior to
SEM analysis a 20 nm chromium layer was deposited by sputtering.
Atomic force microscopy (AFM) studies were conducted using a Keysights (formally Agilent)
5500 AFM. A droplet of bacteria suspension (8x107 CFU/mL) as described in section 2.7 was
applied onto the micropillar substrate and dried at room temperature. Measurements were
carried out in air using intermittent contact mode (tapping mode) utilizing uncoated silicon
9
NCHV cantilevers (Bruker, Santa Clara, CA). These cantilevers have typical resonance
frequencies of 320 kHz and a typical spring constant of 42 N/m (with a tolerance of 20-80
N/m). Due to the pillar size, the scan rate was set to 0.1 Hz and 5 V amplitude was used for
imaging. Height phase-shift images were recorded and line-fitted using PicoView software
supplied by Keysights.
3. Results and discussions
3.1 Substrates decorated with micropillar arrays
We first investigated the wetting and evaporation behavior of water droplets on substrates
(Figure SI-1) decorated with a pillar height of 5 µm (H5), 10 µm (H10) and 15 µm (H15).
After deposition, the wetting transition from Cassie-Baxter state to the Wenzel state24-27 was
clearly visible for substrates H5, H10 and H15 at t ~ 80 sec ± 40 sec. On all our samples the
static CA for water was found to be ~100° (±7°). We measured the CA of the water droplet as
a function of time. The dynamics of CA values of water on these fabricated pillars are displayed
in Figure SI-2. Initial CA was 98o ±6o, 105o ±5o, 100o ±7o for H5, H10 and H15, respectively.
Hysteresis was 20o ± 5, 35o ± 8 and 60o ± 15 for H5, H10 and H15, respectively.43-45 High
hysteresis is expected for wetted surfaces H5, H10 and H15. This caused by a loss on
hydrophobicity followed by droplet impalement in the micropillars. High hysteresis values
have also been observed for polymeric susbtrates.34 It has been reported that capillary forces
applied by sessile droplets can deform elastic surfaces.46 This explains the strong hysteresis we
observe for H15 surfaces in Figures SI-4(c) and (f).
During evaporation, the CA of the droplets decreases (Figure SI-2), zipping-wetting
propagation is observed (shown in Figure SI-5 between t = 800 and 930 sec), which has also
been observed for comparable configurations.1-3 In the previous studies, the zipping-wetting
effect was observed with the propagation of the fluid entering and filling the microstructures
10
as seen in Figure SI-5. The zipping-wetting process of these droplets is energetically favored
at low CA (e.g. t = 650 sec, see Figures SI-2 and SI-5), and it becomes more favorable for the
higher pillars.
3.2 Evaporation of S. epidermidis suspension over substrates with micropillars
In order to investigate the behavior of droplets containing bacteria, three different
concentrations of S. epidermidis suspensions (6.3x106 CFU/mL, 8.0x107 CFU/mL and 5.0x109
CFU/mL) were prepared as described in experimental section 2.5. The pattern of bacterial
distribution after drying is affected by both the concentration of S. epidermidis in the water
droplets, and the height of the pillars as presented in Figure 2. A homogeneous bacterial
distribution is observed for (a) H5, (b) H10 and (c) H15 at the high concentration of bacteria
(5.0x109 CFU/mL). We hypothesize that this cell distribution is governed by a high amount of
S. epidermidis agglomerates at the last moment of evaporation. A microbial adherence test to
n-hexadecane was performed47 to estimate S. epidermidis hydrophobicity. This technique has
been used to qualitatively estimate surface hydrophobicity of cells.48, 49 Cellular interactions
are assumed to be subjected to forces similar to those governing colloidal aggregations between
surfaces or particles in liquid. The hydrophobic interaction forces are strongly attractive and
are determined by the amount of hydrophobic/hydrophilic molecular components on S.
epidermidis (e.g. polysaccharides or hydrophobins). From our experiments, cultured S.
epidermidis cells reveal hydrophobicity of 58% ± 5%. This suggest that attractive forces for
hydrophobic cells interact stronger via Van der Waals forces which could prompt
agglomeration leading to aggregates during evaporation.
As the concentration is reduced to 6.3x106 CFU/mL, the classical ring-shaped stain is not
visible using only white light due to the reduced amount of bacteria. Only few bacterial clusters
at the border of the stain are observed in Figure 2(a)-(c). Moreover, for the intermediated
11
concentration (8.0x107 CFU/mL) an accumulation of bacteria in the center of the octagonal
shape was observed alongside bacterial distribution at the borders (Figure 2(b)). This implies
that the final evaporation patterns depend on a sensitive balance between bacteria and capillary
interactions during the final stages of evaporation. It is important to note that in the current
conditions Marangoni flow is much smaller than the dominant evaporation-driven flow.50, 51
The Zipping-wetting effect was also observed for S. epidermidis containing droplets. Figure 3
shows a top-view image of a droplet containing S. epidermidis deposited over H15. An
irregular octagon was observed until t ~ 700 s, after which the droplet changes into a square
shape, as the fluid fills the cavities between the micropillars. It is observed that at t = 960 s, the
liquid spread out from the corners of the droplet with the formation of a cross structure
stretching outside the square pattern at t =1120 s. Similar effects were also observed for
evaporating droplets with higher bacterial concentration (e.g. 5.0x109 CFU/mL) see supporting
video H15.
To evaluate both, the zipping-wetting and the coffee stain effects during evaporation of droplets
containing S. epidermidis, we studied the distribution of the localized bacterial patterns as well
as bacterial cells viability. An intermediate bacterial concentration of ~ 8.0x107 CFU/mL was
chosen for the work in the following sections as this gave a clear visualization of the dried
bacterial patterns (Figure 2).
3.3 Localized S. epidermidis deposition environments between micropillars
We investigated the localized environment of the bacteria within the troughs of the micropillars
after evaporation using the entire droplet area (i.e. droplet perimeter and center of the droplet).
Figure 4(a) shows a top-view illustration of a square lattice composed by four micropillars
(grey dots) with bacteria (red dots) in the troughs. Different local bacterial environments
between pillars are depicted as follows: a completely filled structure (red box); a square lattice
12
with four filled edges and an empty central space, “O” shape (green box); a three sided
deposition with bacteria in “C” shape (purple box); a two sided “L” shape bacterial distribution
(blue box); and, finally, a single line (“I”) of bacteria (pink box). A top-view bright field
modular microscope image of a micropatterned substrate with deposited S. epidermidis is
shown in Figure 4(b) highlighting the different kinds of local environments that are
experimentally observed for the bacteria. It can be seen that all five environments are observed,
highlighted with an arrow of the same color as used in Figure 4(a). In contrast to the structured
surfaces, flat epoxy surfaces do not contain similar well-defined localized bacterial
configurations. For comparison, a representative image of dried bacteria patterns on a flat
epoxy surface is presented in Figure SI-7.
To establish the detailed distribution of bacteria suggested from the light microscope data,
AFM images were collected. Due to limitation of the depth that can be probed by the AFM,
imaging was only used to identify the deposition of the bacteria on substrate H5 (Figure 1(a)).
The AFM data in Figure 5 shows that a high proportion of S. epidermidis cells were found at
the bottom of the troughs in the space between pillars and a significantly smaller population of
bacteria was found on top of the pillars. AFM images were processed to enhance the contrast
between the floor (purple color), deposited bacteria (light blue colors) and top of pillars (red
color).
AFM image on Figure 5 clearly shows that the deposition shapes observed by light microscopy
in Figure 4(b). This can be directly attributed to the local environment and deposition pattern
of the bacteria (Figure 2(a)). We have therefore mapped the statistical distribution of the
different local environments of the deposited bacteria as the pillar height of the substrate is
changed (shown in Figure 4(c)). It can be seen that the H5 and H10 distribution is comparable,
with a similar distribution for the “O”, “C” and “L” environments (each approximately 15% of
the total number of patterns). In contrast, the H15 has a much higher concentration of
13
completely filled troughs and much fewer low-concentration local environments. H5 and H15
show opposite behavior, with the taller substrate forming high-concentrations of local
environments and vice-versa, while H10 can be considered an intermediate case. Therefore,
the discussion is focused on substrates H5 and H15. Note that proportion of localized S.
epidermidis environments shown in Figure 4(c) are collected from three independent
experiments using the entire area of five dried droplets.
For H5 in Figure 4 (c), the highest proportion of the bacterial environments were found for
clusters in “I” shaped environments (ca. 33% of deposition environments), whereas for H15, a
sevenfold decrease in the proportion of “I” shaped environments is observed. Moreover, there
is an increase in the proportion of totally filled and “O” shaped local environments seen for the
H15 substrate when compared to H5 substrates (37% of the total number of environments for
H15 compared to 14% for H5). These results confirm that the induced bacterial deposition
environments can be tuned by changing the pillar heights. We suggest that the observed
distribution of S. epidermidis in Figure 4(c) can be associated with capillary flow of the
evaporated droplets. Thokchom et al. have reported that motile and nonmotile cells can be
directed with the formation of ring deposits on uncoated substrates.52 Moreover, S. epidermidis
preferential cell attachment to the lower areas between pillar troughs has also been reported22
and we confirm here this observation. This implies that our localized bacterial environments
are actively driven by the flow during evaporation and not by the nonmotile microorganism. It
is important to mention that S. epidermidis configurations may also vary in their size and shape
adapting to the configuration of the decorated surface.
To assess how the local environment affects bacterial growth, bacterial viability of planktonic
cells after rehydration was measured and is shown in Figure 4(d). H5 shows slight bacterial
growth inhibition compared to flat, H10 and H15 substrates. We hypothesize that H5 sample
contains a larger proportion of smaller local environments which could be more vulnerable to
14
dehydration and cell death when compared to the larger local environments which are more
prevalent on the H15 sample.
Biofilm formation assays were performed for S. epidermidis deposited over surfaces.42 This
method provides additional insights on the antibacterial performance of structured materials by
estimating the bacterial biomass formed on surfaces. Here, polystyrene flat surface (PSflat),
flat epoxy surface (Flat), and H5, H10 & H15 epoxy substrates were used. In Figure 6(a), we
present optical density (O.D.) values. Representative images of a well plate for each surface
are also presented. Images were recorded after crystal violet staining for PSflat, Flat, H5, H10
& H15. High levels of S. epidermidis biofilm mass are found for the PSflat substrate with an
O.D. ~ 0.45. A substantial reduction of biofilm mass is obtained for Flat, H5, H10 and H15
epoxy substrates. The lowest O.D. values are ~ 0.12 for Flat and H5, while H10 and H15 are
0.17 and 0.25, respectively. From our biofilm mass optical density assay, measured as intensity
reduction of a light beam transmitted through the biofilm, we have correlate the formed biofilm
mass, measured as total carbon and as cell mass. Biofilm formation assay shows clearly the
importance of both chemical composition of the material and surface topography. It has been
demonstrated that staphylococci show great versatility to adhere to polymers, like polystyrene
materials.53, 54 Thus, when compared to PSflat substrate (i.e. highest biofilm mass), epoxy
surfaces reveal promising material properties which could reduce biofilm mass deposition.
Interestingly, in spite of high S. epidermidis viability in planktonic state (Figure 4(d)), biofilm
formation over epoxy surfaces is relatively impaired. It is clear that S. epidermis viability can
only be affected by the surface topography since no additional surface functionalization was
performed. High levels of viable cells have also been observed for functionalized and non-
functionalized surfaces, whereas the topographic surface remains with fewer bacterial cells.55
To assess the effect of surface topography and its ability to reduce S. epidermidis attachment,
biofilm mass values from Figure 6(a) were normalized to the engineered roughness index
15
(ERI) in Figure 6(b).56, 57 ERI (i.e. ERI = (r x df)/fD) is a dimensionless value used to
characterize surfaces with engineered topographies58 which solely considers the micropillar
geometry, the spatial arrangement of the microstructured substrate, and the size of the
topological features. ERI equation comprises of three parameters, the Wenzel’s roughness
factor (r) which is defined as the ratio of the actual surface area to the projected planar surface
area,59, 60 the depressed surface fraction (fD) as the ratio of the recessed surface area between
the protruded features and the projected planar surface area,58 and the degree of freedom of
movement of the microorganism of the recessed areas (df).56-58
From ERI equation, values for structured substrates were 2.9, 4.8, and 6.7 for H5, H10, and
H15, respectively and the ERI value for flat surfaces (i.e. PSflat and Flat) was 2. Figure 6(b)
shows that PSflat substrate has the highest normalized biofilm mass. Compared to Flat surface,
PSflat has ~75% more formed biofilm mass. Moreover, H5, H10 & H15 substrates shows a
~50% reduction in normalized biofilm mass compare to Flat epoxy substrate. From results in
Figure 6(b), no significant differences are observed between H5, H10 and H15. However, S.
epidermidis attachment to H5, H10 and H15, is observed to be reduced when is normalized to
the geometrical features of the fabricated substrates. Similar trends have been also achieved
when O.D. is normalized to total surface area for H5, H10 and H15. From ERI analysis, beyond
a quantitative assessment, we have obtained understanding of cell-feature interaction which
highlights the importance of the topography on cell attachment.
Two approaches have been used to estimate the antibacterial properties of surfaces. For
evaporated droplets, a small decrease in H5 bacterial viability is observed after rehydration and
planktonic cell colony counting (Figure 4(d)). Compared to PSflat, low level of biofilm formed
on epoxy substrates is observed in Figure 6(a). These results show that, regardless surface
geometry, epoxy surfaces like Flat and H5 have promising antibacterial performance. For
future geometrical designs, H5 substrate has shown the most desirable antibacterial properties
16
capable of reducing bacterial re-growth (Figure 4(d)) and bacterial biomass formation (Figure
6(a)).
3.4 Drying of droplets with PS particles
The bacterial patterns described in previous sections correspond to the last stage of the
deposition process. In order to understand better such deposition patterns, we perform
experiments with PS particles with a mean diameter of 1.2 µm ± 0.04 µm which is comparable
to S. epidermidis cell diameter (0.5 to 1.5 µm). The fluorescent labelling of the PS particles
allows us to observe how the deposition occurs during the evaporation process.
Experiments are performed on substrates H5, H10 and H15. PS particles concentration was 107
particles/mL, which is comparable to the intermediate concentrations used for S. epidermidis
in section 3.3. First instants of the droplet life time are dominated by the zipping-wetting effect,
i.e. the contact line spreads in a step-wise manner through the pillars (e.g. Figure SI-5). As a
consequence of this phenomenon, the droplet perimeter adopts a polygonal shape. As the pillar
height increases from H5 to H15 the corners of the droplet footprint become more squared.
In the last step of the evaporation process, PS particles motion is clearly visible (see supporting
videos). PS particles flow is directed towards the droplet corners. The flow rates increase as
the corners of the droplet contact line become sharper. Surface H5 shows the lowest amount of
PS particles deposits at corners of the droplet perimeter (Figure 7(a)), whereas a higher
concentration of PS particles was seen for the H15 substrate (Figure 7(c)).
Figures 7(a), (b) and (c) are taken from the PS particles supporting videos at the last moment
of evaporation for H5, H10 and H15 substrate, respectively. The PS particles tend to
accumulate in rounded corners close to the contact line as in H5 (Figure 7(a)) with a fewer PS
particles accumulating in the sharper corners for H10 & H15 (Figures 7(b) & (c)). Note that
17
the flow is so strong that in case of Figure 7(b) and (c) the contact line is stretched beyond its
pinning position. Due to the enhanced flow towards the corners, those particles that do not
reach the contact line are distributed along the surface forming an ‘X-shape’. This illustrated
in Figure 7(d).
To quantify the surprising correlation found between the particle accumulation at the corners
and the micropillar height, we measure the fluorescent light intensity emitted by the PS
particles at different locations of the droplet at different time point during evaporation. The aim
is to quantify the particle enrichment at the droplet corners and the depletion at its sides through
the fluorescence light intensity, which is directly proportional to the amount of particles. Note
that the measurements start at 80% of the total evaporation time. At this time the coffee-stain
effect has been already able to drag a large amount of particles to the contact line. Therefore,
all intensity profiles show a sharp increase as r/R approaches 1 (with r the distance to the
contact line and R the droplet radius), i.e. as we reach the droplet's contact line. If we focus our
attention first on the droplet side perimeter, in Figure 8(a) and (c) we see that in both cases
(droplets in H5 and H15, respectively), there is a clear decrease of the light intensity as the time
reaches the final evaporation time (a 50% decrease in H5 and about 75% decrease in H15).
This means that particles are being "removed" from the side of the droplet as the solvent
evaporates. Now we focus on the fluorescent intensity change at the corners of the droplets in
Figures 8(b) and (c) for droplets on H5 and H15, respectively. Here, we clearly observe an
opposite effect: the fluorescence intensity increases in almost 100% from the first time point
measured. This intensity increase at the corners is due to the particle enrichment in the formed
polygonal droplets. Note that despite the sharper corners in H15 (Figure 8(d)), the increase in
intensity is comparable to the H5 case (Figure 8(b)). This is attributed to a large amount of
particles in the H15 travelling beyond the pinning line and go beyond the measurement area
(shown in Figure 7(c)).
18
3.5. Interpretation of the experimental results and physical explanation
In previous sections we have shown a clear correlation between the accumulation of particles
and bacteria at the corners of polygonal droplets. Additionally, surfaces with taller pillars show
a larger deposits accumulating at the corners.
In the first time point measured after droplet deposition on the substrate, the droplet experiences
a wetting transition from a Cassie-Baxter state to a lower-energy Wenzel state by filling the
interspace between the micropillars. Under special geometric conditions and solvents, the
liquid front advances through the pillar array in a step-wise manner known as zipping-wetting,
that gives the polygonal shape to the droplet's perimeter. It is well-known that surfaces with
taller micropillars present sharper corners.1-3 The reason is connected with the smaller
curvature that the liquid menisci are able to adopt when the pillars are higher. The contact line
remains pinned for practically entire process.
In sessile droplets, the evaporation occurs preferentially at the contact line6 and consequently
a capillary flow develops and transports liquid and particles to the droplet's perimeter. Such
flow drags the particles or bacteria towards the perimeter, explaining the high fraction found
at the borders of the droplet. This phenomenon, known as the 'coffee-stain effect' explains the
ring-shaped stains formed by the evaporation of a suspension droplet on flat substrates.
The evaporative flux (J) at the droplet's surface depends on the distance from the contact line
r. For the case of very thin droplets the flux takes the form J(r) ~ DCs/R(r/R)-0.5, where D is the
vapor diffusivity, Cs is the vapor concentration difference, R is the droplet radius and r is a
radial distance from the contact line.
The evaporation process changes dramatically when the contact line curves develop "angular
regions" as described by Popov and Witten.61 They analyzed an idealized case of a perfectly
19
sharp corner (curvature radius Rc = 0 in Scheme 1). They demonstrated analytically that the
evaporative flux near an angular region is strongly enhanced with respect to a straight contact
line. This is expressed as J ~ DCs/R(r/R)-0.7 for an angular wedge of angle α = 90°. Here, we
estimate the outer length scale to be the size of the drop. Therefore, a particle in an evaporating
square-shaped droplet feels a preferential flow towards the corners (see Figure 8). The angular
region at the corner of the droplet is smoothened on a scale r~Rc, i.e. the curvature is not
apparent when one sits very close to the corner. At such a scale, we should recover the square
root behavior J ~ DCs/Rc(r/Rc)-0.5, but now with Rc as the relevant scale.
Assuming that the flow velocity is directly proportional to the evaporative flux6 J, we compare
the flow towards the corners against the flow towards the straight contact line regions. Then,
we can conclude that there is a flow enhancement towards the corners by a factor (R/Rc)1/2,
that in our case is of the order of 10 for the sharpest droplets. Consequently, the smaller the
contact line curvature radius Rc is, the larger its influence in the generated flow towards the
corners. Note that the smallest Rc that can be achieved is limited by the diameter of the smallest
microstructure holding the contact line. In this particular case, the pillars have typical diameters
of 5 µm (therefore Rc=5 µm), while the droplets have typical radius, R, of 1 mm.
4. Conclusions
The evaporation of induced bacterial patterns over micropillared substrates was studied.
Variations in the shape of the deposition patterns are achieved by changing the pillar height of
the fabricated micropatterns. We show that the non-axisymmetric evaporation process is found
to be responsible for the inhomogeneous deposition of particles along the droplets perimeter.
This is a result of the combined action of the coffee-stain effect and the zipping-wetting effect
which results in the breakdown of symmetry of the perimeter of the droplet. Variations in
bacterial distribution are explained by the enhanced evaporation-induced flow towards the
20
corners of the polygonal droplets on the substrates. We observed a sharp difference in the type
of local environment, as the pillar height is increased. The H15 substrates induce the deposition
of bacteria into environments with high local concentration of cells. On the other hand, on the
smaller pillar heights a lower local concentration environment is favored. Our results indicate
that low height microstructured surfaces can lower bacterial regrowth and biomass attachment.
These findings could be utilized for the design of novel topographical antimicrobial surfaces.
21
5. References
1. Sbragaglia, M.; Peters, A. M.; Pirat, C.; Borkent, B. M.; Lammertink, R. G. H.; Wessling,
M.; Lohse, D. Spontaneous breakdown of superhydrophobicity, Phys. Rev. Lett. 2007, 99,
156001
2. Pirat, C.; Sbragaglia, M.; Peters, A. M.; Borkent, B. M.; Lammertink, R. G. H.; Wessling,
M.; Lohse, D. Multiple time scale dynamic in the breakdown of superhydrophobicity, EPL
2008, 81, 66002
3. Peters, A. M.; Pirat, C.; Sbragaglia, M.; Borkent, B. M.; Wessling, M.; Lohse, D.;
Lammertink, R. G. H. Cassie-Baxter to Wenzel state wetting transition: scaling of the front
velocity, Eur. Phys. J. E 2009, 29, 391-397
4. Courbin, L.; Denieul, E.; Dressaire, E.; Roper, M.; Ajdari, A.; Stone, H. A. Imbibition by
polygonal spreading on microdecorated surfaces, Nature Materials 2007, 6, 661-664
5. Raj, R.; Adera, S.; Enright, R.; Wang, E. N.; High-resolution liquid patterns via three-
dimensional droplet shape control, Nature Comm. 2014, 5, 4975
6. Deegan, R. D.; Bakajin, O.; Dupont, T.; Huber, G.; Nagel, S.; Witten, T. Capillary flow as
the cause of ring stains from dried liquid drops, Nature 1997, 389, 827-829
7. Tani, M.; Kawano, R.; Kamiya, K.; Okumura, K. Towards combinatorial mixing devices
without any pumps by open-capillary channels: fundamentals and applications, Scientific
Reports 2015, 5, 10263
8. Fox, M.; Esveld, E.; Luttge, R.; Remko, B. A new pulsed electric field microreactor:
comparison between the laboratory and microtechnology scale, Lab Chip 2005, 5, 943–948
9. Lemée, F.; Clarot, I.; Ronin, L.; Aranda, L.; Mourer, M.; Regnouf-de-Vains, J.-B. A
bacteriophilic resin, synthesis and E. coli sequestration study, New J. Chem. 2015, 39,
2123-2129
22
10. Scheuerman, T. R.; Camper, A. K.; Hamilton, M. A. Effects of substratum topography on
bacteria adhesion, Journal of colloids and interface science 1998, 208, 23-33
11. Merian, T.; Goddard, J. M.; Advances in nonfouling materials: perspectives for the food
industry, Journal of Agricultural and Food Chemistry 2012, 60, 2943-2957
12. Crawford, R. J.; Webb, H. K.; Truong, V. K.; Hasan, J.; Ivanova, E. P. Surface
topographical factors influencing bacterial attachment, Advances in Colloids and Interface
Science 2012, 179-182, 142-149
13. Duncan, T. V.; Applications of nanotechnology in food packaging and food safety: Barrier
materials, antimicrobials and sensors, Journal of Colloid and Interface Science 2011, 363,
1-24
14. Desrousseaux, C.; Sautou, V.; Descamps, S.; Traor, O. Modification of the surfaces of
medical devices to prevent microbial adhesion and biofilm formation, Journal of Hospital
Infection 2013, 85, 87-93
15. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from
January 1992 through June 2004, issued October 2004. Am. J Infect Control. 2004, 32,
470–485
16. Renner, L. D.; Weibel, D. B. Physicochemical regulation of biofilm formation, MRS
bulletin 2011, 36, 347-355
17. Hochbaum, A. I.; Aizenberg, J. Bacteria pattern spontaneously on periodic nanostructure
arrays, Nano Lett. 2010, 10, 3717-3721
18. Allion, A.; Baron, J.-P.; Boulange-Petermann, L. Impact of surface energy and roughness
on cell distribution and viability, Biofouling 2006, 22, 269-278
19. Medilanski, E.; Kaufmann, K.; Wick, L. Y.; Wanner, O.; Harms, H. Influence of the cell
surface topography of stainless steel on bacterial adhesion, Biofouling 2002, 18, 193-203
23
20. Bruzaud, J.; Tarrade, J.; Coudreuse, A.; Canette, A.; Herry, J.-M.; de Givenchy, E. T.;
Darmanin, T.; Guittard, F.; Guilbaud, M.; Bellon-Fontaine, M.-N. Flagella but not type IV
pili are involved in the initial adhesion of Pseudomonas aeruginosa PAO1 to hydrophobic
or superhydrophobic surfaces, Colloids and Surfaces B: Biointerfaces 2015, 131, 59-66
21. Xu, L.-C.; Siedlecki, C. A. Staphylococcus epidermidis adhesion on hydrophobic and
hydrophilic textured biomaterials, Biomed. Mater. 2014, 9, 035003
22. Perera-Costa, D.; Morales Bruque, J.; González-Martín, M. L.; Gómez-García, A. C.;
Vallido-Rodríguez, V. Studying the influence of Surface topography on bacterial adhesion
using specially organized microtopographic surface patters, Langmuir 2014, 30, 4633-4641
23. Tsai, P.; Lammertink, R. G. H.; Wessling, M.;. Lohse, D. Evaporation-triggered wetting
transition for water droplets upon hydrophobic microstructures, Physical Review Letters
2010, 104, 116102
24. Papadopoulos, P.; Mammen, L.; Deng, X.; Vollmer, D.; Butt, H.-J. How
superhydrophobicity breaks down, Proc. Natl. Acad. Sci. USA 2013, 110, 3254–3258
25. Dietrich, S.; Popescu, M. N.; Rauscher, M.; Wetting on structured substrates, J. Phys.:
Condens. Matter 2005, 17, S577–S593
26. Bonn, D.; Eggers, J.; Indekeu, J.; Meunier, J.; Rolley, E. Wetting and spreading, Rev. Mod.
Phys. 2009, 81, 739-805
27. Susarrey-Arce, A.; Marin, A. G.; Schlautmann, S.; Lefferts, L.; Gardeniers, J. G. E.; van
Houselt, A. One-step sculpting of silicon microstructures from pillars to needles for water
and oil repelling surfaces, J. Micromech. Microeng. 2013, 23, 025004
28. Marin, A. G.; Gelderblom, H.; Susarrey-Arce, A.; van Houselt, A.; Lefferts, L.; Gardeniers,
J.G.E.; Lohse, D.; Snoeijer, J. H. Building microscopic soccer balls with evaporating
colloidal fakir drops, Proc. Natl. Acad. Sci. USA 2012, 109, 16455–16458
24
29. Barbieri, L.; Wagner, E.; Hoffmann P.; Water Wetting transition parameters of
perfluorinated substrates with periodically distributed flat-top microscale obstacles,
Langmuir 2007, 23, 1723–1734
30. Guoping, F.; Amirfazli, A. Understanding the edge effect in wetting: a thermodynamic
approach, Langmuir 2012, 28, 9421-9430
31. Susarrey-Arce, A.; Marín, A. G.; Nair, H.; Lefferts, L.; Gardeniers, J. G. E.; Lohse, D.; van
Houselt, A. Absence of an evaporation-driven wetting transition on omniphobic surfaces,
Soft Matter 2012, 8, 9765-9770
32. do Nascimento, R. M.; Cottin-Bizonne, C.; Pirat, C.; Ramos S. M. M. Water Drop
Evaporation on Mushroom-like Superhydrophobic Surfaces: Temperature Effects,
Langmuir 2016, 32, 2005-2009
33. Reyssat, M.; Yeomans, J. M.; Quéré, D. Impalement of fakir drops, EPL 2008, 81, 26006
34. Chuang, Y.-C.; Chu, C.-K.; Lin, S.-Y., Chen, L.-J. Evaporation of water droplets on soft
patterned surfaces, Soft Matter 2014, 10, 3394-3403
35. Boreyko, J. B.; Baker, C. H., Poley, C. R., Chen, C.-H. Wetting and Dewetting Transitions
on Hierarchical Superhydrophobic Surfaces, Langmuir 2011, 27, 7502-7509
36. Xie, X.; Li, Y.; Zhang, T.; Fang, H. H. P. Bacterial survival in evaporating deposited
droplets on a teflon-coated surface, Appl. Microbiol Biotechnol. 2006, 73, 703-712
37. Nellimoottil, T. T.; Rao, P. N.; Ghosh, S. S.; Chattopadhyay, A. Evaporation-induced
patterns from droplets containing motile and nonmotile bacteria, Langmuir 2007, 23, 8655-
8658
38. Jerrim, L. B.; Velev, O. D. Deposition of coating from live cells and large particles by
“convective-sedimentation” assembly, Langmuir 2009, 25, 5692-5702
39. Baughman, K. F.; Maier, R. M.; Norris, T. A.; Beam, B. M.; Mudalige, A.; Pemberton, J.
E.; Curry, J. E. Evaporative deposition patterns of bacteria from sessile drop: effect of
25
changes in surface wettability due to exposure to a laboratory atmosphere, Langmuir 2010,
26, 7293-7298
40. Sempels, W.; De Dier, R.; Mizuno, H.; Hofkens, J.; Vermant, J. Auto-production of
biosurfactants reverses the coffee ring effect in a bacterial system, Nature Comm. 2013, 4,
1757
41. Marin, A.; Gelderblom, H.; Lohse, D.; Snoeijer, J. H. Order-to-disorder transition in ring-
shaped colloidal stains, Phys. Rev. Lett. 2011, 107, 085502
42. O'Toole, G. A. Microtiter dish biofilm formation Assay, J. Vis. Exp. 2011, 47, e2437
43. de Gennes, P.-G.; Brochard-Wyart, F.; Quéré, D. Capillarity and Wetting Phenomena:
Drops, Bubbles, Pearls, Waves, Springer-Verlag, New York, 2004, ISBN 0-387-00592-7
44. Reyssat, M.; Quéré, D. Contact angle hysteresis generated by strong dilute defects, J. Phys.
Chem. B 2009, 113, 3906–3909
45. Yeh, Y.-H.; Cho, K.-H.; Chen, L.-J. Effect of softness of polydimethylsiloxane on the
hydrophobicity of pillar-like patterned surfaces, Soft Matter 2012, 8, 1079-1086
46. Pericet-Camara, R.; Auernhammer, G. K.; Koynov, K.; Lorenzoni, S.; Raiteric, R.;
Bonaccurso, E.; Soft Matter 2009, 5, 3611-3617
47. Krepsky, N.; Barreto, R.; Ferreira, R.; Nunes, A. P. F.; Lins, U. G. C.; Filho, F. C. S.; de
Mattos-Guaraldi, A. L.; Netto-dosSantos, K. R. Cell surface hydrophobicity and slime
production of Staphylococcus epidermidis Brazilian isolates, Current Microbiology 2003,
46, 280-286
48. Rosenberg, M.; Gutnick, D.; Rosenberg, E. Adherence of bacteria to hydrocarbons: a
simple method for measuring cell-surface hydrophobicity, FEMS Microbiology Letters
1980, 9, 29-33
49. Israelachvili, J. N.; McGuiggan, P. M. Forces between surfaces in liquids, Science 1988,
241, 795-800
26
50. Marin, A.; Liepelt, R.; Rossi, M.; Kähler C. J. Surfactant-driven flow transitions in
evaporating droplets, Soft Matter 2016, 12, 1593-1600
51. Hernández-Sánchez, J. F.; Eddi, A.; Snoeijer, J. H. Marangoni spreading due to a localized
alcohol supply on a thin water film, Phys. Fluids 2015, 27, 032003
52. Thokchom, A. K.; Swaminathan, R.; Singh, A. Fluid flow and particle dynamics inside an
evaporating droplet containing live bacteria displaying chemotaxis, Langmuir 2014, 30,
12144-12153
53. Otto, M., Staphylococcal Biofilms, Curr Top Microbiol Immunol. 2008, 322, 207-228
54. Heilmann, C.; Hussain, M.; Peters, G.; Götz, F., Evidence for autolysin-mediated primary
attachment of Staphylococcus epidermidis to a polystyrene surface, Molecular
Microbiology 1997, 24, 1013-1024
55. Susarrey-Arce, A.; Sorzabal-Bellido, I.; Oknianska, A.; McBride, F.; Beckett, A. J.;
Gardeniers, J. G. E.; Raval, R.; Tiggelaar, R.; Diaz Fernandez, Y. A. Bacterial viability on
chemically modified silicon nanowire arrays, J. Mat. Chem. B 2016, 4, 3104-3112
56. Graham, M. V.; Cady, N. C. Nano and microscale topographies for the prevention of
bacterial surface fouling, Coatings 2014, 4, 37-59
57. Graham, M. V.; Mosier, A. P.; Kiehl, T. R.; Kaloyeros, A. E.; Cady, N. C. Development of
antifouling surfaces to reduce bacterial attachment, Soft Matter 2013, 9, 6235-6244
58. Schumacher, J. F.; Carman, M. L.; Estes, T. G.; Feinberg, A. W.; Wilson, L. H.; Callow,
M. E.; Callow, J. A.; Finlay, J. A.; Brennan, A. B. Engineered antifouling
microtopographies-effect of feature size, geometry, and roughness on settlement of
zoospores of the green alga Ulva, Biofouling 2007, 23, 55-62
59. Wenzel, R. N. Resistance of solid surfaces to wetting by water, Ind. Eng. Chem. 1936, 28,
988-994
60. Bico, J.; Tordeux, C.; Quere, D. Rough wetting, Europhys. Lett. 2001, 55, 214-220
27
61. Popov, Y. O.; Witten, T. A. Characteristic angles in the wetting of an angular region:
deposit growth, Physical Review E 2003, 68, 036306
Acknowledgements
We acknowledge Dr. Marco Marcello and Dr. Joanna Wnetrzak from Liverpool Centre for Cell
Imaging (CCI) for provision of imaging equipment and technical assistance as well as
Nanoinvestigation Centre for access to the facility. The authors also express their gratitude to
S. Schlautmann, M. Bos and G. W. Bruggert from University of Twente for technical support.
This work was partly founded by EPSRC grant number EP/J019364/1.
28
Figures and captions
Figure 1. SEM images of substrates with micropillars: (a) 5 µm height (H5), (b) 10 µm height
(H10), and (c) 15 µm height (H15).
(a) H5
10 μm 10 μm 10 μm
Sid
e-vi
ew S
EM
imag
e
10 µm
(b) H10 (c) H15
h d i
29
Figure 2. Images of the resulting patterns formed after the evaporation of S. epidermidis
droplets containing 5.0x109 CFU/mL (first row), 8.0x107 CFU/mL (second row) and 6.3x106
CFU/mL (third row) over (a) H5, (b) H10 and (c) H15 substrates. For all substrates, the scale
bars in the light microscope images represent 250 μm. In addition, S. epidermidis stain outside
of the original square pattern and is highlighted in red in (b).
(a) H5 (b) H10
250 µm
5x109 CFU/ml
8x107 CFU/ml
6x106 CFU/ml
(c) H15
30
Figure 3. Top-view images of a droplet containing ~8x107 CFU/mL S. epidermidis deposited
and evaporated over H15 surface. Direction of zipping-wetting effect is highlighted with a red
arrow. The scale bar at the bottom right represents 250 μm.
S. epidermidisH15
t = 50 sec t = 640 sec t = 700 sec t = 710 sec
t = 830 sec t = 960 sec t = 1100 sec t = 1120 sec
31
Figure 4. (a) Sketch of drying patterns of evaporated droplets with S. epidermidis between
micropillar troughs. From left to right: completely filled structure, square lattice with empty
central space, bacteria in “C” shape, bacteria in “L” shape and “I” single line of bacteria. (b)
Representative bright field modular microscope image of an evaporated droplet area over H5
containing S. epidermidis patterns. Highlights represent a bacterial environment for each
category identified by color in (a-b). (c) Chart of the percentage of S. epidermidis patterns
deposited in H5, H10 and H15. (d) Count number of viable S. epidermidis cells recovered after
24 h after rehydration on flat surface and on substrates decorated with micropillars H5, H10
and H15. Experiments in (c) were performed in triplicates by drying ten to twenty independent
droplets over substrates. The number of pattern in (c) was estimated from five entire evaporated
droplets per dried substrate. Microbiological test in (d) were carried out independently in
triplicates. Values in (c) and (d) were expressed ±SD.
0
10
20
30
40
50
H5
0
10
20
30
40
50
H10
0
10
20
30
40
50
H15
(c)
Flat
H5
H10
H15
108
109
1010
1011
1012
CFU mL-1 cm
-2
Pill
ar
he
igh
t
(d)
(a)
50 µm
(b)
(b)
(a)
(c)
50 µm
S. epidermidis
(d)
108
109
1010
1011
1012
Ba
cte
rial via
bili
ty o
f S
. e
pid
erm
idis
incu
ba
ted
ove
r ep
oxy s
urf
aces (
CF
U/m
L)
Flat H5 H10 H15
%
32
Figure 5. 3D-AFM image of a H5 surface with S. epidermidis patterns deposited at the bottom
of the troughs and atop of pillars. Patterns formed by S. epidermidis are highlighted with
colored arrows as shown in Figures 4(a) and 4(b). Note that from 3D-AFM image, the lower
plane between the micropillars troughs is purple and bacteria on the floor of the surface are in
blue colors.
10 µm
H5
S. epidermidisatop of a pillar
Atop of pillars
Pillar sides
Pillar sides
S. epidermidis
S. epidermidis
S. epidermidis
Bottom
33
Figure 6. (a) Biofilm formation assay with S. epidermidis cultured for 24h over surfaces:
polystyrene flat (PSflat), flat epoxy (Flat), H5, H10 and H15. (b) Normalized biofilm mass to
ERI for PSflat, Flat, H5, H10 and H15. Three independent experiments were performed. All
values are expressed ±SD.
0.00
0.05
0.10
0.15
0.20
0.25
0.30
No
rmaliz
ed
bio
film
mass to E
RI
Flat H5 H10 H15PSflat
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Bio
film
ma
ss (
O.D
. 5
95
nm
)
Flat H5 H10 H15PSflat
(a) (b)
34
Figure 7. Drying patterns from evaporated droplets containing PS particles on (a) H5, (b) H10
and (c) H15 substrates. (d) Preferential direction drawing of the capillary-driven flow is
highlighted with blue arrows. In addition, fluorescent particles stretching outside of the original
square patterns (see (b)-(c)) are highlighted with an open dashed circle. Preferential direction
of the capillary driven flow contributing to the distribution of the particles is also highlighted
with an arrow.
35
Figure 8. Fluorescent light intensity emitted by PS particles. The intensity is proportional to
the particle density. Measurements in (a, c) and (b, d) were performed during drying of a droplet
over substrate H5 & H15, respectively. (a) & (c) shows the intensity change from the center to
the side perimeter of the droplet, while (b) and (d) show the intensity change from the center
of the droplet to the corner. Intensity measurements are presented during last intervals before
complete evaporation, e.g. 80% (black line), 90% (red line), and 98% (blue line).
0.0 0.2 0.4 0.6 0.8 1.00
1000
2000
3000
4000
5000
6000
7000
8000
Inte
nsit
y c
ou
nts
(A
rb. U
nit
is) t/tfinal = 80%
t/tfinal = 90%
t/tfinal = 98%
r/R
0.0 0.2 0.4 0.6 0.8 1.00
1000
2000
3000
4000
5000
6000
r/R
Inte
nsit
y c
ou
nts
(A
rb. U
nit
is) t/tfinal = 80%
t/tfinal = 90%
t/tfinal = 98%
0.0 0.2 0.4 0.6 0.8 1.00
500
1000
1500
2000
2500
3000
3500
4000
t/tfinal = 80%
t/tfinal = 90%
t/tfinal = 98%
Inte
ns
ity
co
un
ts (
Arb
. U
nit
is)
r/R
0.0 0.2 0.4 0.6 0.8 1.00
1000
2000
3000
4000
5000
6000
7000
8000
9000
t/tfinal = 80%
t/tfinal = 90%
t/tfinal = 98%
r/R
Inte
nsit
y c
ou
nts
(A
rb. U
nit
is)
(a) H5 (b) H5
(c) H15 (d) H15
PS particles
corner rrside
perimeter
corner rrside
perimeter
36
Scheme 1. (a) Side view of a deposited droplet on a substrate with a sharpness curvature and
contact angle (CA) in r-z planes (b) Top-view of a droplet with geometrical curvature in r-Φ,
Rc is the corner’s radius of curvature and α is the wedge angle (c) Detail of the droplet corner:
r is defined as the distance to the contact line and J is the evaporative flux.
37
Tables
Table 1. Height (h), pillar-to-pillar interspace (i) and diameter (d) of the microstructures on
substrates (a)-(c)
Microstructure h (μm) i (μm) d (μm)
(a) H5 4.8 4.7 5.0
(b) H10 9.5 4.5 5.0
(c) H15 15.7 5.0 5.2
38
For table of content (TOC) use only
Pattern formation by S. epidermidis
on micropillars arrays Drying of droplets with PS particles
vs.